The use of levonorgestrel-releasing intrauterine system in prevention of endometrial pathology in women with breast cancer treated with tamoxifen.
Abstract: Objective

To determine whether the use of levonorgestrel-releasing intrauterine system prevents endometrial pathology in women with breast cancer treated with tamoxifen or not.

Participants and methods

We did a randomized controlled trial on 121 women who required adjuvant therapy with tamoxifen for breast cancer. The women were randomly assigned to either group A (endometrial surveillance alone) or group B (endometrial surveillance before and after insertion of the levonorgestrel intrauterine system for two years). Endometrial surveillance was done by outpatient hysteroscopy and endometrial biopsy before and two years after the start of tamoxifen.


The baseline assessment showed only benign uterine changes in all women (n=121).Women in group B had a much lower incidence of endometrial polyps than group A (1.8 % versus 16.1 % p value = 0.02). There was no statistical significant difference in the incidence of submucous fibroid between the two groups (6.4% in group A, 3.3% in group B , P value = 1.1). There was a higher incidence of bleeding in group B but this resolved to a baseline similar to that of group A.


The levonorgestrel-releasing intrauterine system has a protective action against the endometrial pathology caused by tamoxifen therapy in women with breast cancer. It reduces the occurrence of de novo endometrial polyps.

Keywords: levonorgestrel-releasing intrauterine system, tamoxifen, breast cancer, endometrial polyps.
Article Type: Report
Subject: Levonorgestrel (Usage)
Levonorgestrel (Health aspects)
Breast cancer (Drug therapy)
Breast cancer (Research)
Intrauterine contraceptives (Usage)
Tamoxifen (Usage)
Tamoxifen (Health aspects)
Authors: Omar, Hassan
Elkhayat, Waleed
Aboulkasem, Mohamed
Pub Date: 01/01/2010
Publication: Name: Archives: The International Journal of Medicine Publisher: Renaissance Medical Publishing Audience: Academic Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2010 Renaissance Medical Publishing ISSN: 1791-4000
Issue: Date: Jan-March, 2010 Source Volume: 3 Source Issue: 1
Topic: Event Code: 310 Science & research
Product: Product Code: 3842296 Intrauterine Devices NAICS Code: 339113 Surgical Appliance and Supplies Manufacturing SIC Code: 3842 Surgical appliances and supplies
Geographic: Geographic Scope: Egypt Geographic Code: 7EGYP Egypt
Accession Number: 226568397

Tamoxifen a nonsteroidal triphenylethylene is the most commonly used hormonal treatment for breast cancer. It is highly effective in lowering the risk of recurrent or contralateral breast cancer regardless of menopausal status or clinical stage of disease. Tamoxifen is not a pure antioestrogen, since it has an oestrogenic effect in skeletal muscles, in lipid metabolism and in various gynaecological tissues. This unopposed estrogenic effect promotes occult uterine lesions to develop into polyps, fibroids and endometrial hyperplasia and causes a 2-3 fold increase in endometrial cancer. (1,2) The incidence of endometrial polyps, hyperplasia and endometrial cancer has been reported to be between 5 and 35%, 4.7-16% and 0.8-5% respectively. (3-5) These lesions commonly cause bleeding episodes which although benign in most cases require investigations to exclude malignant disease. These episodes compromise the quality of life of women taking tamoxifen and have big implications for health resources. Routine endometrial screening of women without symptoms on tamoxifen has been suggested but would not be cost effective and would be unlikely to lower mortality from endometrial cancer. (6) An alternative strategy to deal with tamoxifen--induced endometrial changes and the symptoms they cause is to render the uterine tissues unresponsive to estrogenic stimulation using progestagen. The use of systemic progestagens although established in prevention of the endometrial neoplastic changes associated with oestrogen replacement therapy, unfortunately causes several unwanted side effects leading to poor compliance and discontinuation of the treatment. There is also some concern that high dose systemic progestagens may blunt the efficacy of tamoxifen to prevent recurrence of breast cancer. (7) In addition some authors reported that systemic progestagens did not reverse the development of polyps, cysts and fibroids associated with tamoxifen. (8) Levonorgestrel-releasing intrauterine system delivers a high dose of progestagens in the uterine tissues with a low dose systemically thus decreasing the progestagenic side effects and has also been shown to induce regression of endometrial hyperplasia. (9)

In this study, we evaluate the efficacy of the levonorgestrel-releasing intrauterine system in preventing endometrial pathology in women with breast cancer treated with tamoxifen.


Study design

We did a randomised controlled trial in Kasr El Aini Hospital in the period from May 2006 until June 2009. Women who consented to participate in the study were randomised to the LNG-IUS treatment or control group according to a computer generated random number series in serially numbered sealed envelopes. The follow up period was two years after inclusion in the study. The study was approved by the ethical committee of the department of Obstetrics & Gynecology, Cairo University.


The inclusion criteria were:

Women with early stage breast cancer requiring postoperative adjuvant tamoxifen were eligible after completion of postoperative radio and chemotherapy, no history of progestagen treatment since the diagnosis of breast cancer, no known allergy to polyethylene , healthy pelvic organs by physical examination and ultrasonography, satisfactory hysteroscopic assessment with all surfaces of the uterine cavity clearly visualized, uterine cavity less than 8 cm measured by transvaginal ultrasound, and age less than 60 years. All patients gave written consent prior to the study.

The exclusion criteria were: pelvic inflammatory disease, active liver disease, history of malignant disease other than breast cancer and refusal of levonorgestrel intrauterine system. The baseline characteristics including age, parity, menopausal status, body mass index, age at diagnosis, stage of the disease, and adjuvant therapy were recorded.

The intrauterine system:

We used the levonorgestrel intrauterine system (Mirena, Schering Health Care, West Sussex, UK). It releases 0.02mg of levonorgestrel per day producing a very high local concentration in the endometrial tissue with low plasma concentrations. (10)

Study protocol

At the start, a baseline endometrial surveillance was done with transvaginal ultrasonography to determine the uterine size, endometrial thickness, and to exclude uterine and adnexal pathology. Endometrial thickness was defined as the widest point across the hyperechoic area in a longitudinal section of the uterus along the sagittal plane. The length of the uterine cavity was defined as distance between the internal cervical os and the most distant extreme of the hyperechoic area. Follow up ultrasound measurements were done after one and two years (end of the study) without knowledge of previous measurements to ensure objectivity. We did an outpatient hysteroscopic assessment with endometrial sampling at the start of the study and after two years of tamoxifen treatment (end of the study) with a 3 mm rigid hysteroscope with a normal saline as the distending medium. The endometrial biopsy is taken from the uterine fundus and the posterior uterine wall. Any suspicious lesion, polyp or fibroid should be removed and subjected to histopathological examination. All specimens were fixed with hematoxylin and eosin and examined with a pathologist who was blinded to the randomizations. The women randomised to the treatment group had their LNG-IUS inserted immediately after the baseline hysteroscopy. Women were advised to report any potential side effects of the device; headache, lower abdominal and back pain, nausea, oedema, mastalgia, mood changes, vaginal dryness and abnormal vaginal bleeding. The women in the treatment group had their device removed two years after fitting at the time of follow-up hysteroscopy and biopsy sample assessment.

Statistical analysis

Data were analysed using SPSS version 13.0 (SPSS Inc., Chicago, IL, USA) statistical program. Chi-square test or Fisher's exact test was used to examine group differences as an appropriate at an intention- to-treat basis. A probability value less than 0.05 was considered significant.


There were 150 women eligible for the study (81 premenopausal and 69 postmenopausal). After randomization eight women from the control group and ten women from the treatment group were declined to participate. During baseline hysteroscopy, the procedure was unsuccessful in four women in the control group and five women in the treatment group. After one year of tamoxifen treatment, a woman in the control group died from breast cancer and a woman in the treatment group suffered from menometrorrhagia and had undergone hysterectomy. Finally, 121 women (62 in the control group and 59 in the treatment group) completed the two-year follow-up period. There is no statistical significant difference between the two groups regarding the baseline characteristics (Table1).

Hysteroscopic assessment before the start of tamoxifen treatment, we found 6 women in the control group and 3 women in the treatment group had asymptomatic endometrial polyps. The difference was not statistically significant (P = 0.4) (Table 2). All polyps were less than 15 mm in diameter. The polyps were resected hysteroscopically and all were benign on histological examination. No submucosal fibroids were found at the baseline hysteroscopy. Histological assessment of the endometrial samples at the baseline showed no statistical significant difference between the control and the treatment group (Table 2). Hysteroscopic assessment at two years (end of the study), we found 10 women in the control group and one women in the treatment group, had asymptomatic endometrial polyps. The difference between both groups was statistically significant (16.1% in the control group versus 1.8% in the treatment group. P = 0.02) (Table 3). All polyps were less than 15 mm in diameter. The polyps were resected hysteroscopically and all were benign on histopathology. 4 women in the control group and two women in the treatment group had submucosal fibroids. All fibroids were less than 3 cm in diameter. The difference was not statistically significant (P = 1.1) (Table 3). Histological assessment of endometrial biopsy at two years, there were histological patterns in the control group similar to those of the baseline assessment. In the treatment group, most of the cases (96.6) showed atrophic or inactive endometrium (Table 3). Abnormal vaginal bleeding was reported at one year by two women in the control group and 22 women in the treatment group which was statistically significant (P <0.001) (Table 4). The majority of bleeding occurred in the first 6 months. At two years abnormal vaginal bleeding was reported by two women in the control group and 7 women in the treatment group. The difference between both groups was not statistically significant (P =0.08) (Table 4).


Tamoxifen has been found as an antioestrogen to significantly reduce the risk of recurrent or contralateral breast cancer. Tamoxifen has been associated with benign changes of the endometrium including cystic atrophy and polypoid changes. Endometrial polyps are rarely associated with neoplastic changes, with only a 0.5% chance for cancer present within the polyp. (11) In our study we found that prophylactic use of LNG-IUS before the start of tamoxifen treatment effectively reduced its oestrogenic effect on the endometrium. There was a nine-fold reduction in the incidence of endometrial polyps at two years(the duration of the study), this finding was similar to that of Chan et al and Gardner et al. (12,13) There was no case of endometrial hyperplasia or atypia in both groups at two years. Over 96% of the cases in the treatment group had either atrophic or inactive endometrium. Proliferative or secretory endometrium occurred in over 16% in the control group but none in the treatment group. All endometrial polyps in our study were benign however malignancy within a polyp has been associated with tamoxifen use. (14) Therefore, endometrial polyps in tamoxifen-treated women even if asymptomatic should be removed. As regard the incidence and size of fibroids, it has been suggested that the use LNGIUS may reduce the size of a fibroid in tamoxifen-treated women (13) however in our study no significant difference neither in the number nor in the size of the fibroids between both groups. Adverse effects of LNG-IUS such as mastalgia, acne or headache were uncommon. The only significant adverse effect of the intrauterine system was an abnormal vaginal bleeding. Such bleeding may be simply a foreign body effect or a hormonal effect although a hormonal effect is more likely because women fitted with a standard intrauterine device rarely have such long- term bleeding. Abnormal vaginal bleeding was more in the treatment group than the control group, especially in the first 6 months, after this time the bleeding patterns in the treatment group approached those of the baseline patterns in the control group. This pattern was comparable to that reported by Chan et al. (12)


In conclusion, the levonorgestrel-releasing intrauterine system has a protective action against the endometrial pathology caused by tamoxifen therapy in women with breast cancer. It reduces the occurrence of de novo endometrial polyps. To make for a routine use of the device for each woman prescribed tamoxifen, there is a need for a large randomised controlled trail to include the insertion of the device for the standard duration of tamoxifen treatment, which is five years.

Conflict of interest: None declared.


(1.) Van Leeuwen FE, Benraardt J, Coebergh JW, Keimeney LA. Risk of endometrial cancer after tamoxifen treatment of breast cancer. Lancet 1994;343:448-52.

(2.) Assikis VJ, Neven P, Jordan VC, Vergote I. A realistic clinical perspective of tamoxifen and endometrial carcinogenesis. Eur J Cancer. 1996;32A:1464-76.

(3.) Neven P, De Muylder X, Van Belle Y, Van Hooff I, Vanderick G. Logitudinal hysteroscopic follow up during tamoxifen treatment. Lancet 1998;351:36.

(4.) Vosse M, Renard F, Cobion M, Neven P, Ngaret JM, Hertens D. Endometrial disorders in 406 breast cancer patients on tamoxifen: the case for less intensive monitoring. Eur J Obstet Gynecol Reprod Biol. 2002;101:58-63.

(5.) Ozsener S, Ozaran A, Ltil L, Dikmen Y. Endometrial pathology of 104 postmenopausal breast cancer patients treated with tamoxifen. Eur J Gynecol Oncol. 1998;19:580-3.

(6.) Neven P, Vergote L. Should tamoxifen users be screened for endometrial lesions? Lancet 1998; 351:155-57.

(7.) Panay N, Studd J. Progestogen intolerance and compliance with hormonal replacement therapy in menopausal women. Hum Reprod Update. 1997;3: 159-71.

(8.) Powles TJ, Bourne T, Athanassiou S, Chang J, Grubock K, Ashely S. The effects of norethisterone on endometrial abnormalities identified by transvaginal ultrasound screening of healthy postmenopausal women on tamoxifen or placebo.Br J Cancer. 1998; 78:272-75.

(9.) Wildemeersch D, Dhont M .Treatment of nonatypical and atypical endometrial hyperplasia with a levonorgestrel-releasing intrauterine system. Am J Obstet Gynecol 2003;188:1297-8.

(10.) Nilsson CG, Haukkamaa M, Vierola H. Luukkainen T. Tissue concentrations of levonorgestrel in women using levonorgestrel releasing IUD. Clin Endocrinol. 1982;17:529-36.

(11.) Ramondetta LM, Sherwood JB, Dunton CJ, Palazzo JP. Endometrial cancer in polyps associated with tamoxifen use. Am Jobstet Gynecol. 1999;180:340-1.

(12.) Chan SS, Tam WH, Yeo W, Yu MM, Ng DP, Wong AW, Kwan WH, Yuen PM. A randomized controlled trial of prophylactic levonorgestrel intrauterine system in tamoxifen-treated women BJOG. 2007;114:1510-1515.

(13.) Gardner FJE, Konje JC, Abrams KR, Brown LJR, Khana S, et al.Endometrial protection from tamoxifen-stimulated changes by a levonorgesrel-releasing intrauterine system: a randomized controlled trial. Lancet. 2000;356:1711-17.

(14.) Ismail SM. Pathology of endometrium treated with tamoxifen. J Clin Pathol. 1994;47:827-33.

Corresponding author: Waleed Elkhayat, M.D., 5 Qura Ibn Shourik Street, Giza, 12211, Egypt

Hassan Omar, Waleed Elkhayat, Mohamed Aboulkasem

Department of Obstetrics & Gynecology, Faculty of Medicine, Cairo University, Cairo, Egypt
Table 1. Baseline characteristics of the 121 participants who
completed the two year follow up period (data are presented
as mean (SD) or n (%).

                     Control Group   Treatment        P Value
                     (n = 62)        Group n (=59)

Mean age (years)     51.5(8.9)       47.9(8.0)        0.07

Nulliparity          8(12.9)         6(9.1)           0.6

women                28(45.1)        21(35.5)         0.28

BMI (Kg/[m.sup.2])   26.9(4.7)       26.5(4.9)        0.89

Age at diagnosis     48.3(7.3)       45.1(8.6)        0.79
of breast cancer

Stage of disease

Stage 1              22(35.4)        16(27.1)         0.27
Stage 2              38(61.2)        40(67.7)         0.8
Stage 3              2(3.2)          3(5.08)          0.61

Adjuvant             34(54.8)        37(62)           0.76

Adjuvant             36(58)          35(59.3)         0.87

Table 2. Hysteroscopic and histological findings before
tamoxifen treatment.

                       Control Group   Treatment        P Value
                       (n = 62)        Group n (=59)
fi ndings

Polyps                 6(9.7)          3(5)             0.4

Submucous fibroids     0               0                --

Histology of
endometrail biopsy

for diagnosis          17(27.4)        13(22)           0.32

Atrophic or inactive   31(50)          29(49.1)         0.39

or secretory           14(22)          17(28.8)         0.57

Hyperplasia            0               0                --

Table 3. Hysteroscopic and histological finding after two
year tamoxifen treatment.

                        Control Group   Treatment        P Value
                        (n = 62)        Group (n=59)


Polyps                  10(16.1)        1(1.8)           0.02

Submucous fibroids      4(6. 4)         2(3 .3)          1.1

Histology of
endometrial biopsy

for diagnosis           15(24.2)        2 (3.4)          <0.001

Atrophic or inactive    37(59.6)        57(96.6)         <0.001

or secretory            10(16.2)        0                <0.001

Hyperplasia             0               0                --

Table 4. Vaginal bleeding in both groups after one and after
two years of follow-up.

                   Control Group   Treatment        P Value
                   (n = 62)        Group n (=59)

Abnormal           2(3.2)          22(37.2)         <0.001
vaginal bleeding
after one year

Abnormal vaginal   2(3.2)          7(11.8)           0.08
bleeding after
two years
Gale Copyright: Copyright 2010 Gale, Cengage Learning. All rights reserved.